Exploring the antioxidant activity of Fe(III), Mn(III)Mn(II), and Cu(II) compounds in Saccharomyces cerevisiae and Galleria mellonella models of study

Abstract Reactive oxygen species (ROS) are closely related to oxidative stress, aging, and the onset of human diseases. To mitigate ROS-induced damages, extensive research has focused on examining the antioxidative attributes of various synthetic/natural substances. Coordination compounds serving as synthetic antioxidants have emerged as a promising approach to attenuate ROS toxicity. Herein, we investigated the antioxidant potential of a series of Fe(III) (1), Mn(III)Mn(II) (2) and Cu(II) (3) coordination compounds synthesized with the ligand N-(2-hydroxybenzyl)-N-(2-pyridylmethyl)[(3-chloro)(2-hydroxy)]-propylamine in Saccharomyces cerevisiae exposed to oxidative stress. We also assessed the antioxidant potential of these complexes in the alternative model of study, Galleria mellonella. DPPH analysis indicated that these complexes presented moderate antioxidant activity. However, treating Saccharomyces cerevisiae with 1, 2 and 3 increased the tolerance against oxidative stress and extended yeast lifespan. The treatment of yeast cells with these complexes decreased lipid peroxidation and catalase activity in stressed cells, whilst no change in SOD activity was observed. Moreover, these complexes induced the Hsp104 expression. In G. mellonella, complex administration extended larval survival under H2O2 stress and did not affect the insect's life cycle. Our results suggest that the antioxidant potential exhibited by these complexes could be further explored to mitigate various oxidative stress-related disorders.


Introduction
Oxidativ e str ess is c har acterized by the excessiv e pr oduction of oxidizing a gents, suc h as r eactiv e oxygen (ROS) and nitr ogen (RNS) species, whic h ar e counter acted by the pr esence of antioxidants .T he disruption of redox balance is multifactorial, and it has been related to endogenous (e.g.mitochondrial function and the activity of xanthine, cytoc hr ome P450, monoamine and NADPH oxidases) and exogenous (e.g.exposure to tobacco, medications, UV and/or ionizing radiation) factors (Sies 2015 ;Poprac et al.2017 , During o xidati v e str ess, ROS-induced r eactions may cause dr amatic c hanges in v arious biological structur es that ar e curr entl y associated with the incidence of various o xidati ve stressr elated diseases, suc h as cancer and neur odegener ativ e diseases (e.g.Alzheimer´s and Parkinson´s disease) (Valko et al. 2007, Sies 2015, Poprac et al. 2017 ;Tan et al. 2018 .In this scenario, it is imper ativ e to de v elop ne w str ategies to mitigate ROS r eactivity and the o xidati v e dama ges in proteins, lipids and nucleic acids. All aer obic or ganisms hav e e volv ed a sophisticated antioxidant system, composed of enzymatic components [e.g.superoxide dismutases (SODs), catalases (CATs), glutathione peroxidases (GPXs) and per oxir edoxins (PRXs), methionine sulfoxide reductase (MsrA), 8-Oxoguanine DNA Glycosylase (OGG1) and mutY DNA glycosylase (MUTYH)], as wells as non-enzymatic biomolecules [e.g.glutathione (GSH), vitamins C, E and A, and thioredoxin (TRx)].These elements play a crucial role in ROS elimination or redox signaling (Benzie 2000, Valko et al. 2007, Birben et al. 2012 ).The perfect functioning of this antioxidant system is critical for an effectiv e cellular r esponse a gainst R OS.Among the w ell-kno wn enzymatic antioxidant r epr esentativ es , GPXs , C ATs and SODs metalloenzymes are considered the primary antioxidant barrier protecting cells from the high reactivity of R OS (Greenw ald 1990 , Birben et al. 2012 , Ighodaro and Akinlo y e 2018 , Sepasi Tehrani and Moosa vi-Mo vahedi 2018 ).It should be noted that these enzymes share a common feature; they all have a metal center in their active site.In the GPx enzyme, selenium is r epr esented by the selenium deri vati ve of c ysteine, selenoc ysteine (Sec), while C ATs ha ve the pr otopor phyrin IX gr oup containing ir on(III) or a dinuclear active site of manganese.SODs may present different metals (e.g.iron, manganese, copper or nickel) at their active sites (Sarma and Mugesh 2008, Batinic-Haberle et al. 2014, Sepasi Tehrani and Moosa vi-Mo vahedi 2018 ).T hese metal ions pla y a k e y role in the enzymatic catalysis of these enzymes: while SOD isoforms use their redox metal centers at the active site to promote the dis-mutation of O 2 r − to O 2 and H 2 O 2 , the GPx and CATs use similar r edox r eactions to dispr oportionate H 2 O 2 into H 2 O and O 2 (Sarma and Mugesh 2008, Batinic-Haberle et al. 2014, Sepasi Tehrani and Moosa vi-Mo v ahedi 2018 ).Notabl y, these enzymes hav e been r eported as potential ther a peutic candidates for reducing primary o xidati v e e v ents c har acterized by an incr ease in ROS and the consequent attack on biomolecules, as well as secondary e v ents r elated to the redox regulation of various processes, such as inflammation (Miao andSt. Clair 2009 , Nandi et al. 2019 ).
Similar to natural enzymes, the catalytic reduction of R OS b y synthetic antioxidants can be ac hie v ed using coordination compounds (Day 2009, Batini ć-Haberle et al. 2010 ).Amongst the sever al adv anta ges of using suc h compounds, whic h include r educed size, high cell permeability, increased circulating half-life, low antigenicity, and reduced manufacturing costs), the most pr omising pr operties observ ed in this class of compounds are their catalytic efficiency in the elimination of H 2 O 2 and/or O 2 − and their capacity to modulate cellular redox environment (Day 2009, Batinic-Haberle et al. 2014 ).Metalloporphyrins, macrocyclic N-containing ligands and salen deri vati ves are the most notorious r epr esentativ es of antioxidant compounds, and promising efficacy in attenuating oxidative stress has been reported in several models of study (e.g. in vitro and experimental animals), including for human diseases (Tian et al. 1993, Li et al. 2003, Doctrow et al. 2012, Bonetta 2018 ).Due to the r ele v ance and a pplicability of such a class of compounds, many other metal-based compounds have been described as synthetic antioxidants.Our group reported a series of coordination compounds synthesized with the ligand 1-[bis(pyridin-2-ylmethyl)amino]-3-c hlor opr opan-2-ol (HP-ClNOL) that sho w ed promising catalytic activity in vitro , and in Sacc harom yces cerevisiae was able to reduce H 2 O 2 , O 2 r -and HO r toxicity and, consequentl y, incr ease cell surviv al under o xidati ve stress and ageing (Ribeiro et al. 2015, 2017, Menezes et al. 2023 ).
Inter estingl y, the efficacy of these complexes appears to be related to the reduction of protein and lipid oxidation, as well as the modulation of metal homeostasis in the intracellular milieu (Ribeiro et al. 2015(Ribeiro et al. , 2017 ) ). Mor e r ecentl y, a 1,10-phenantr oline-octanediaoate Mn 2 + -complex was reported exhibiting significant antioxidant potential, modulating intracellular oxidation and decreasing the susceptibility of S. cerevisiae cells and G. mellonella larvae to oxidativ e str ess (Queir oz et al. 2022 ).Mor eov er, this Mn 2 + -complex also atten uated the to xicity and a ggr egation of alpha-synuclein ( αSyn), a human protein related to Parkinson´s disease, by interacting with negativ el y c har ged acidic r esidues (e.g.Asp135) located at the C-terminal region of αSyn monomers (Queiroz et al. 2022 ).Altogether, it is somewhat evident that synthetic antioxidants become an inter esting alternativ e for ther a peutic use a gainst oxidativ e str ess-r elated pathologies.
Her ein, we inv estigated the potential of a series of a Fe(III), Mn(II)Mn(III) and Cu(II) coordination compounds synthesized with the N 3 O donor ligand H 2 BPClNOL, N-(2-hydroxybenzyl)-N-(2pyridylmethyl)[(3-c hlor o)(2-hydr oxy)]-pr opylamine, in pr otecting Sacc harom yces cerevisiae , whic h was singl y tr eated with complexes and stressed with H 2 O 2 and menadione, an O 2 r − radical gener- ator.Additionall y, we anal yzed the efficacy of these compounds in extending the longevity of yeast cells using a yeast c hr onolog ical ag ing model.This study focused on e v aluating c hanges in cell damage as well as the activation of primary (e.g.activation of SOD and CAT) and secondary (e.g.expression and activation of heat shoc k pr otein 104-Hsp104) cellular responses after treatment with the complexes.Given its involvement in the yeast response to various stressors, including o xidati ve stress, we investigated whether the singular treatment of cells with the com-plexes could independently induce the expression of Hsp104.Finall y, we explor ed the benefits of tr eatment with the complexes in the larvae of G. mellonella , an inv ertebr ate model of study.The larv ae wer e subjected to H 2 O 2 str ess, and we addr essed the impact of the treatment with the complexes on the insect's life cycle.(H 2 BPClNOL) (Horn Jr. et al. 2000 ) and its metal-based deri vati ves [Fe(HBPClNOL) (Fernandes et al. 2010 ) were synthesized as previously reported (Fig. 1 ).Infrared and elemental (CHN) anal yses wer e carried out and the r esults a gr ee with the data reported in the literature.

Antioxidant activity in vitro
For the analysis of the complex's antio xidant acti vity, the in vitro micr oscale DPPH r adical (2,2-diphen yl-1-picrylhydr azyl) ca ptur e method was used (Mensor et al. 2001 ).DPPH samples initially prepared to 250 μM in ethanol were then successively diluted in the same solvent to a final concentration of 3.91, 7.82, 15.63, 31.25, 62.5, 125, and 250 μM.To avoid the interference of colors associated with the compounds, a blank system containing the complexes diluted in ethanol, at eac h concentr ation tested was also pr epar ed.The auto reduction of DPPH was used as a control to e v aluate the antioxidant percentage of the complexes.After serial dilution and controls preparation, 50 μL of DPPH was added to each well, in a 96-well microplate, except for the blank system pr epar ed to r educe the interfer ence r elated to the color of the complexes.Samples were k e pt protected from light at room temper atur e and after 30 min of reaction, the absorbance was measured at 518 nm.To calculate the RSA 50, the GraphPad Prism 9.0.0 software was used.Data were normalized, and the tool "Asymmetric Sigmoidal, 5PL, X is log(concentration) " was applied (Chen et al. 2013 ).
The interaction between complexes and H 2 O 2 or O 2 r − was also investigated spectroscopically by UV-Vis and EPR.Stock solutions of the complexes (1 mM) were prepared in PBS (0.1% DMSO).For UV-Vis spectroscopy, 540 μL of the complex solution was added to a quartz cuvette and its spectrum was obtained.
Then, 60 μL of a 1 M solution of H 2 O 2 was added to the cuvette (ratio complex:H 2 O 2 ∼ 1:100) and the spectra were recorded every 12 s for 5 min.For EPR studies, H 2 O 2 (150 μL, 100 mM) was added to an EPR tube containing 150 μL of the complex to be anal yzed.The r atio complex:H 2 O 2 w as1:100.Readings w ere carried at 120 K after 2, 30 and 60 min.To e v aluate the interaction between complexes and O 2 r − , stock solutions (1 mM) for each complex and KO 2 (2 and 5 mM) were prepared in DMSO.The initial spectra of each complex were recorded by UV-Vis and EPR.
In UV-Vis, aliquots of 10 μL of KO 2 (2 mM) were added to the cuvette and spectra were recorded after each addition.The reaction was also monitor ed thr ough time, with the addition of a single aliquot of 100 μL of KO 2 to the cuvette and spectra were r ecorded e v ery 12 s for 5 min.For EPR studies, differ ent volumes of KO 2 solution were added to an EPR tube containing 150 μL of the complex stock solution.The volume was completed to 300 μL with DMSO and readings were carried out after 2 min reaction at 120 K.
Oxidati v e stress tolerance and the adapti v e treatments with 1, 2, or 3 Exponentiall y fermentativ e cells wer e dir ectl y exposed to oxidizing agents (1 mM H 2 O 2 or 30 mM menadione) and incubated at 28 • C and 160 rpm for 1 h (Pereira et al. 2003, Castro et al. 2007 ).
To analyze the antioxidant activity of 1, 2, and 3 , the cells were pr etr eated with incr easing concentr ations of the complexes for 1 h at 28 • C and 160 rpm (Ribeiro et al. 2015, Queiroz et al. 2022 ).
Next, the cells wer e harv ested, washed twice with sterile water, resuspended in the original culture medium, previously centrifuged, and then subjected to o xidati v e str ess.Toler ance was anal yzed by plating treated and non-treated cells in 2% solid YPD medium, in triplicate, before and after o xidati ve stress.Control cells (nontreated and non-stressed) were used as the control for tolerance determination, calculated as the ratio between the number of colonies forming units counted after o xidati v e str ess and the control condition.

Detection of lipid per o xidation
Lipid peroxidation was determined by the TBARS (thiobarbituric acid r eactiv e substances) method, whic h detects the final pr oduct of lipid oxidation, malondialdehyde (MDA), as pr e viousl y described (Steels et al. 1994 ).

Determination of catalase (CAT) and super o xide dismutase (SOD) activities
Enzyme activities were carried out in cell-fr ee extr act submitted or not to treatment with the complexes (Mariani et al. 2008 ).Activities were also determined in samples treated with the complexes and then exposed to o xidati v e str esses (H 2 O 2 or menadione).CAT activity was performed by enzymatic kinetics following the pr ogr ess of H 2 O 2 consumption at 240 nm (Aebi 1984 ).Catalase activity was expressed as μmol/min/mg ptn, in which one catalase unit is defined as the amount of the enzyme that cat-al yzes the degr adation of 1.0 μmol H 2 O 2 per min ute.Sod acti vity was performed by monitoring the inhibition of NBT reduction (Gamero-Sandemetrio et al. 2013 ).After the zymogram, the polyacrylamide gel electrophoresis was digitalized on EC3 UVP bioimaging system, and SOD activity was determined using Ima geJ softwar e, whic h considers the ar ea of SOD band density.The activity was expressed as a fold increase in SOD activity, calculated by the ratio of complex treated or H 2 O 2 -stressed and control (non-treated and non-stressed) cells.

Expression and activ a tion of the heat shock protein, Hsp104
The expression and activation of the Hsp104 protein was monitored by fluorescence microscopy in the BY4741 S. cerevisiae str ain geneticall y modified to expr ess the GFP-ta gged Hsp104 (Kr osc hwald et al. 2015 ).The expression and activation of the Hsp104 protein were investigated in cells submitted or not to a 1 h treatment with the complexes 1, 2, and 3.As a positive control, the expression and activation of the Hsp104 protein was assessed by submitting yeast cells to a mild heat-shoc k tr eatment at 40ºC/1 h.The expression of Hsp104 was determined by calculating the percentage of fluorescent cells in relation to the untreated cells .T he activation of Hsp104 was determined by calculating the percentage of cells with foci formation, referring to Hsp104 assembling with protein aggregates.

Chronological lifespan and complexes adapti v e treatment
Chr onological a ging was determined in post-mitotic yeast cells, accordingl y (Subr amaniyan et al. 2019 ).P ost-mitotic (3-da y stationary phase) cells were chronologically aged for 28 days at 160 rpm/28ºC in the absence or in the presence of 1, 2, or 3 in the cultur e medium.Chr onological lifespan was determined by plating aged cells every 4 da ys .Untreated cells plated on day 0 of the experiment (3-day stationary phase cells) were considered 100% alive.
The treatment with 1-3 was also performed by direct injection of complexes.Ho w e v er, in this case, complexes wer e administer ed in the right-left pro-leg into the hemocoel of G. mellonella larvae 3 h before the H 2 O 2 injection.All experiments were performed in triplicate and each group was formed by 10 larvae.As a con- trol for mechanical injury, in all experiments, a group of 10 larv ae r eceiv ed a dir ect injection with 10 μl of sterile distilled water, as described abo ve .Survival was monitored at 24 h intervals until the end of the insect`s life cycle.Larvae that did not respond to stimulation performed by touch using a micropipette tip were considered dead.

Sta tistical anal ysis
All experiment data were analyzed using Graph Pad Prism 9 (San Diego, California, USA) softwar e.All v alues wer e expr essed as mean values ± standard deviation (SD) of at least three independent experiments and were analyzed using one-way Anova or tw o-w a y Ano v a , whic h denotes homogeneity between experimental groups at P < 0.05.

Results
The in vitro antioxidant activity of 1, 2, and 3 It was reported previously that 1-3 show SOD and CAT-like activity.Ho w e v er, amongst these complexes, 3 sho w ed such activity only in the presence of a co-catalyst (piperazine) (Costa et al. 2018, Guerr eir o et al. 2020 ).Her ein 1-3 wer e e v aluated a gainst the DPPH radical.In comparison, the w ell-kno wn antioxidant complex, [Mn(SALEN)(Cl)] (EUK-8), w as emplo y ed as a r efer ence to e v aluate the r adical scav enging activity a gainst DPPH.As shown in Fig. 2 , the DPPH radical scavenging activity of the tested complexes increased dose-dependently.The DPPH radical was completel y scav enged by the ir on compound (1) at a concentr ation of 125 μM while the same result was r eac hed by 2 only at 250 μM.Compound 3 decomposed around 80% of the radical at the highest concentration.Regarding the DPPH radical scavenging activity of [Mn(SALEN)(Cl)], our results sho w ed that the profile presented b y this standar d antioxidant w as resembled to that observed for the complex 3.The RSA 50 (Radical Scavenging Activity 50: concentration of the complex capable of reducing DPPH by 50%) of each complex was e v aluated.As shown in Table 1 , 2 pr esented the lo w er RSA 50 (30.8± 0.02 μM) in comparison to 1 (64.2 ± 0.01), 3 (67.1 ± 0.03) and [Mn(SALEN)(Cl)] (64.9 ± 0.06).This result demonstrates that although 2 r eac hed 100% antio xidant acti vity only at 250 μM concentr ation, contr asting to that observed for 1, the r equir ed concentration of 2 to reduce DPPH by 50% was lower than that presented by the other complexes.Of r ele v ance is the fact that 64.9 ± 0.06 The calculated RSA 50 values of complexes were obtained from a non-linear r egr ession curv e dose ver sus antio xidant acti vity determined by the DPPH method.RSA 50 values of complexes were obtained as the mean ± standard deviation of at least 3 inde pendent experiments.* Re pr esents statisticall y differ ent r esults in r elation to [Mn(SALEN)(Cl)] ( P < 0.05).
compound 2 exhibited superior RSA 50 compared to [Mn(Salen)Cl], while compounds 1 and 3 demonstrated a similar RSA 50 to the selected standard antioxidant compound.

Complexes act as antioxidants protecting S. cerevisiae cells from oxidati v e stress
To further e v aluate the in vivo antio xidant acti vity of the complexes, the tolerance of the wild type strain of S. cerevisiae , treated or not with 1-3 was analyzed under o xidati v e str ess gener ated by H 2 O 2 (1,0 mM/1 h) and menadione (30 mM/1 h) ( Fig. S1 ).Before assessing the antioxidant capacity of the complexes, it was verified whether S. cerevisiae cells would be susceptible to high concentrations of these complexes.In contrast to the chosen conditions of o xidati v e str ess, tr eating cells with complexes did not impair the survival of S. cerevisiae cells ( Fig. S2 ), indicating that the complexes exhibited non-toxic properties to w ar ds y east cells.
Regarding the antioxidant potential of these complexes, all compounds sho w ed a similar pattern of cell protection.Accor ding to Fig. 3 , we can observe that the treatment with the lo w est concentration (6.25 μM) of the tested complexes was already able to protect yeast cells against the o xidati ve stress generated by H 2 O 2 .The survival of cells directly exposed to 1.0 mM of H 2 O 2 for 1 h was 16.9%, and after pr e vious tr eatments with 6.25 μM of 1-3, follo w ed b y H 2 O 2 stress, a significant rise in the survival rates (47.5%, 41.4% and 36.4%,r espectiv el y) was observed (Fig. 3 A-C).The most significant cellular pr otection a gainst exposur e to H 2 O 2 was ac hie v ed under tr eatment employing 25 μM and 50 μM of the complexes.Ho w e v er, compound 3, at a concentration of 12.5 μM, could also confer cell protection at the same rates observed for the treatments with 25 μM and 50 μM (Fig. 3 C).Unexpectedl y, tr eating cells with 100 μM of all tested complexes decr eased cell pr otection compar ed to the group of cells treated with 25 and 50 μM.Ho w e v er, the surviv al of cells treated with 100 μM was still higher than those dir ectl y str essed with H 2 O 2 (1 mM).
Since all the complexes protected yeast cells against H 2 O 2 stress, we further investigated whether these complexes would also be able to protect cells of the wild type and its isogenic mutant deficient in the synthesis of the Cu,Zn-Sod ( sod1 ) from the str ess gener ated by menadione (30 mM/1 h), an O 2 r − gener ating drug.After menadione stress, cells of the wild type strain were dr asticall y affected, pr esenting onl y 2.0% surviv al under O 2 r − stress (Fig. 4 A-C).The treatment with 1 or 3 increased the survival of the wild type strain in a dose-dependent manner (Fig. 4 A and C).On the other hand, the treatment with 2 presented a distinct profile and did not protect the wild type cells up to the treatment with 25 μM; ho w e v er, after tr eatment with 50 μM and 100 μM, we observed a significant increase in yeast protection when compared to the untreated cells (Fig. 4 B).As expected, the lack of Cu,Zn-

Complexes treatment reduces lipid oxidation and differentially modulates endogenous SOD and CAT antioxidant activities
The increase in cell survival after o xidati ve stress may result from reducing o xidati ve damage and/or eliminating R OS b y treatment with 1-3.Then, we next investigated whether treatment with the complexes would be able to attenuate the lipid peroxidation promoted by stress with H 2 O 2 .As shown in Fig. 5 , it is possible to observe that o xidati ve stress caused a drastic increase in lipid oxidation (i.e. a 3.4-fold increase) compared to the non-stressful condition.After treatment with the complexes, a significant decrease in the le v els of lipid oxidation could be observed; ho w ever, this reduction was not able to r estor e the lipid oxidation le v els observ ed in unstressed cells (Fig. 5 ).Moreover, we did not observe any differ ences r egarding the ca pacity of complexes to r educe yeast lipid oxidation under H 2 O 2 stress.All complexes were able to reduce lipid oxidation to the same magnitude.Next, the activities of CAT and SOD wer e inv estigated to assess whether complex treatment could modulate the activities of these ROS-scavenging enzymes.As expected, cells undergoing fermentative metabolism sho w ed lo w le v els of CAT activity.In contrast, after submitting the cells to o xidati ve stress (1 mM H 2 O 2 /1 h), an increase (i.e.3.2 times) in CAT activity was observed (Fig. 6 A and B).Treating the cells with the complexes did not increase CAT activity at the tested concentrations (25 μM and 50 μM) (Fig. 6 A).Inter estingl y, after tr eatments with the complexes and subsequent exposure to H 2 O 2 stress, we observed that CAT activity presented a similar trend that was found for the unstressed cells (control) (Fig. 6 A).This result might suggest that the 1-3 might deal with the stressor agent (H 2 O 2 ), avoiding the r equir ement of CAT activation.
Regar ding SOD activity, w e observ ed a 2.7-fold incr ease in the activity of this antioxidant enzyme after H 2 O 2 stress when compared to unstressed cells (Fig. 6 B).Unlikely to C AT profile , the SOD activity did not change after singly treatments with the tested complexes, and the SOD activity, after 1-3 treatments follo w ed b y H 2 O 2 str ess r emained as high as observ ed in H 2 O 2 -stressed cells (Fig. 6 B)

Singl y trea tments with 1-3 induced Hsp104 expression, albeit no changes in its activity were observed
To determine the expression of the Hsp104 protein after treatment with the tested complexes, the BY4741 strain was genetically modified to express the Hsp104 fused to the green fluorescent protein, GFP (Hsp104-GFP).This construction allo w ed us to analyze the Hsp104 expression (i.e. increase in fluorescence) and activ ation (i.e.incr ease in foci formation) after singly treatments with the complexes.According to Fig. 7 , we can see that under normal growth conditions, no fluorescence signal emitted by the Hsp104-GFP construct was observed.This result shows that under our experimental condition the Hsp104 expression is not inducible and is not r equir ed under non-stressful conditions (Fig. 7 A and C).As expected, the mild heat stress (40 ºC/1 h) induced an increase in both the expression and the activation of Hsp104 (Fig. 7 A, B and C).Mor eov er, foci formation was distinctl y observ ed in 75% of fluor escent cells.Inter estingl y, the singl y complex tr eatments increased GFP fluorescence, indicating that these complexes promoted the induction of Hsp104 expression in yeast cells.Howe v er, it was observed that the complexes did not activate Hsp104, as evidenced by the absence of foci formation in fluorescence microscopy.

Complexes treatment extends yeast life span
Since all complexes increased yeast tolerance under o xidati ve stress conditions, we further investigated whether the tested complexes could increase yeast lifespan during c hr onological a ging.Lifespan was assessed in c hr onologicall y a ged cells of the wild type strain of S. cerevisiae , BY4741, which was treated or not with the complexes (25, 50 and 100 μM).Chronological aging was monitored for 28 da ys , and every 4 da ys , the cells were collected and plated in 2% YPD medium to assess cell viability.We can see those treatments with the complexes at the highest concentrations (50 or 100 μM) extended yeast lifespan during c hr onological aging (Fig. 8 A-C).After 28 days of chronological aging, the non-tr eated a ged cells sho w ed a cellular lifespan of 32%, whilst the lifespan of cells that were treated with 50 μM or 100 μM of eac h complex significantl y incr eased to 44%-49% and 55%-59%, r espectiv el y (Fig. 8 A-C).Mor eov er, our r esults sho w ed that the antia ging pr ofile of these complexes was dose-dependent since the treatment of cells with the lo w er concentration of complexes was somewhat similar to untreated aged cells.

Complexes decrease G. mellonella larvae susceptibility to H 2 O 2 stress and did not alter the insect life cycle
Finally, the antioxidant potential of the complexes was investigated in the Galleria mellonella model of study, which was subjected to acute o xidati v e str ess by dir ect injection of H 2 O 2 into the larval hemocoel.As shown in Fig. 9 (A, B and C), all H 2 O 2 -stressed larvae died up to the 3 rd day of injection.Susceptibility to H 2 O 2 was partially reverted after treatment with the complexes 1, 2 or 3 (Fig. 9 A, B, and C).T he group of G .mellonella larv ae tr eated with 50 mg kg −1 of complexes sho w ed remarkable survival to H 2 O 2 stress than untreated larvae.Survival increased at 1 and 3 days after treatments with complexes 2 or 3 and 1, r espectiv el y (Fig. 9 A,  B, and C).By increasing the dose to 125 mg kg −1 , we observed that the survival of G. mellonella larvae was improved, reaching 20% survival for 1 and 10% for 2 or 3 on the 7 th day after H 2 O 2 injection.Finall y, tr eatment of larvae with 250 mg kg −1 of 1 or 2 increased larv ae surviv al to 30% on the 7 th day of monitoring, whilst the same treatment with 3 increased larvae survival to 20%.Therefor e, our r esults sho w that the per centa ge of G. mellonella larv ae that surviv ed H 2 O 2 str ess incr eased consider abl y with pr e vious treatment with the antioxidant complexes.As expected, the single administration of 1-3 did not affect larvae survival, having no impact on the insect's life cycle ( Fig. S7 ).
The differentiation process from larva to moth that occurred during the insect's life c ycle w as also monitored after administration of H 2 O 2 , with and without treatment with the complexes.-C) over a 7-day period and depict the ov er all de v elopment thr oughout the entir e insect life cycle (9D-F).We can observ e the thr ee sta ges: larv ae (D), pupae (E), and moth (F).The gr a phs tr ac k the continuity of the animal's life cycle ov er 21 da ys .Results r epr esent the mean ± standard de viation of at least thr ee independent experiments.The demonstration that complexes could not interfere with the G. mellonella life cycle was also crucial to confirm that the treatment with these complexes was safe for the insect.As shown in Fig. 9 , all the larvae of the control group reached the last stage of the insect´s life cycle, and after 20 days of follow-up, the larvae that had turned into pupa, between 6th and 14th da ys , differentiated into a moth.Since all larvae exposed to o xidati v e str ess died on the third day of exposure, it was impossible to assess the interference of H 2 O 2 in the Galleria mellonella life cycle.In contrast, all larv ae tr eated with the highest dose of complexes (250 mg.kg −1 ) and survived until the 7 days of H 2 O 2 stress were able to differen-tiate into a pupa between the 8th and 11th day of life.From the 13th day of life, the pupa gives rise to the adult insect, the moth (Fig. 9 D-F).

Discussion
The antioxidant potential of natural or synthetic substances is c har acterized by their effectiv e ability to eliminate or, at least, atten uate the reacti vity of R OS. Ho w e v er, substances capable of modulating the activity of endogenous pr otectiv e factors and/or reducing o xidati ve damage can also be considered antioxidants of r ele v ance.Her ein, examining the antioxidant potential of three complexes containing iron (1), manganese (2) and copper (3) synthesized with the ligand N-(2-hydroxybenzyl)-N-(2pyridylmethyl)[(3-c hlor o)(2-hydr oxy)]-pr opylamine (H 2 BPClNOL), w e sho w ed that all complexes protected S. cerevisiae and G. mellonella from o xidati ve stress.
Further than the SOD and CAT-like activities pr e viousl y r eported for the compounds ( Figs S3 -S6 ) (Costa et al. 2018 ), we investigated whether the complexes would have some discernible antio xidant acti vity on the DPPH radical.A significant radical scavenging activity (RSA) was observed after incubating DPPH with all tested complexes .T he EC 50 v alue r e v ealed that (2) pr esented the lo w est EC 50 compar ed to the other complexes, whic h is almost 50% lo w er than the result presented by 1 and 3. Considering that compound 2 has two metal ions (Mn) and also two molecules of the ligand while the others have just one, the best activity displayed by 2 may be due to its higher metal and ligand amount.Ther efor e, considering the concentr ation of metal and ligand molecules, the compounds sho w ed similar RSA activities on DPPH.
Mor eov er , employing UV -Vis and EPR spectroscop y, w e could understand the interaction of the complexes 1 and 3 with the H 2 O 2 and O 2 − r , the str essor a gents used in this study.Such investigation has already been reported for compound 2 (Costa et al. 2018 ).Spectr oscopicall y, compounds 1 and 3 show different behavior in the presence of H 2 O 2 .Both UV-Vis and EPR indicate that compound 1 undergoes reduction in the presence of H 2 O 2 since the LMCT phenolateáFe(III) disa ppear ed and the EPR signal associated with the Fe(III) center decreased significantly ( Figs S3 and S4 ).In contr ov ersy, the spectr al featur es of the copper compounds did not change in the presence of H 2 O 2 .
Concerning the interaction of 1 with O 2 − r the UV-Vis indicates possible coordination of the superoxide to the iron compound.The EPR also supports this inter pr etation since at a ratio of 1:1 ), the signal associated with Fe(III) becomes more symmetric and at a ratio of 1:2 it disa ppears, whic h indicates the reduction of the Fe(III) to Fe(II) ( \ Figs S4 and S5 ).The EPR data related to the interaction of 3 with O 2 − r also support that a redox process is taking place since the signal associated with the Cu(II) decreased with the increase in the O 2 − r concentration, indicating the formation of Cu(I).Furthermor e, a ne w pattern at g ∼2.0 has emerged at a ratio 3 :superoxide 1:4, which is attributed to the intermediated Cu(I)-O 2 r − ( Fig. S5 ) (Menezes et al. 2023 ).UV-Vis spectroscop y sho w ed a new band at 450 nm, attributed to a LMCT O 2 r − áCu(II) ( Fig. S4 ) (Fujisawa et al. 1994, Abe et al. 2019, Quek et al. 2021, Menezes et al. 2023 ).
We have previously reported the antioxidant potential of coordination compounds containing the same metal ions (F e , Mn and Cu) but synthesized with different ligand 1-[bis(pyridin-2ylmethyl) amino]-3-c hlor opr opan-2-ol (HPClNOL) using the yeast S. cerevisiae model of study (Ribeiro et al. 2015, 2017, Thornton et al. 2016 ).We described that besides having in vitro antioxidant capacity, all the complexes investigated also pr otected Sacc harom yces cerevisiae from acute H 2 O 2 and O 2 r − stresses.In our previous stud- ies (Ribeiro et al. 2015 ), the antioxidant protection follows the order Fe > Cu > Mn complexes.In contrast, in this work, the copper complex sho w ed the best pr otection.Of r ele v ance is the fact that 3 efficiently protected the sod1 mutant strain, which lacks the Cu,Zn-SOD (SOD1) antioxidant enzyme, when exposed to o xidati v e str ess .T his r esult is v ery inter esting since 3 seems to functionall y r e place the antio xidant enzyme Cu,Zn-SOD, rescuing the high susceptibility phenotype of this mutant strain.Mor eov er, this pr otectiv e pr ofile was some what similar to that ob-tained for the complexes containing the HPClNOL ligand series.In this context, it is worth mentioning that se v er al synthetic antioxidants, including [Mn(Salen)Cl], failed to r e v ert sod1 deficiency [38], and amongst all synthetic antioxidants tested (MnSalen, Mnmacr ocyclic Mn-por phyrin deriv ativ es) onl y MnTM-2-PyP was effective to rescue sod1 defects [38].
High le v els of ROS ar e intimatel y r elated to the accum ulation of o xidati v e dama ge, whic h leads to the onset of a ging and a ger elated diseases (e.g.neur odegener ativ e diseases and cancer) (Cadenas andDavies 2000 , Tan et al. 2018 ).Ther efor e, ther e has been a growing interest in antioxidant compounds to attenuate the effect of o xidati v e str ess and a ging.In our study, we observ ed that all compounds extended yeast lifespan, e v en though no statistical differences were observed between these complexes.Similar antia ging r esults wer e obtained by testing other synthetic antioxidants (Ribeiro et al. 2015(Ribeiro et al. , 2017 ) ).In this context, although we did not observe any differences between the complexes reported her e, our r esults seem pr omising since the treatment of C. elegans with the synthetic antioxidants EUK-8 and EUK-134, capable of increasing the endogenous SOD activity did not extend the nematode lifespan (Keaney et al. 2004 ).
Examining the activity of SOD and CAT antioxidant enzymes, the expression of HSP104 and the lipid peroxidation profile, we observed that the compounds might also be involved in cellular r epr ogr amming to maintain intr acellular r edox status.Activation of antioxidant enzymes is a common mechanism for the acquisition of resistance to o xidati ve stress.In S. cerevisiae , it has been reported that de novo synthesis of antioxidant enzymes increases under ROS stimuli (Moye-Rowley 2002, Herr er o et al. 2008 ).Amongst the transcription factors involved in the response of S. cerevisiae to stress conditions, the transcription factors Yap1 and Hsf1 are the major sensors and signal transducers that regulate gene expression in response to o xidati ve stress (Morano et al. 2012, Mejía-Barajas et al. 2017, Rodrigues-Pousada et al. 2019 ).T hus , acti vating antio xidant enzymes and/or heat shock proteins (HSP´s) is critical for the acquisition of resistance against o xidati ve stress (Per eir a et al. 2001b(Per eir a et al. , 2003 ) ). Contrasting with the results reported in the liter atur e for [Mn(Salen)Cl] (K eaney et al. 2004 ), we did not detect any increase in SOD and/or CAT activity after the treatment with the complexes.Inter estingl y, catalase activity was reduced after treating yeast cells with the complexes and then subjected to o xidati v e str ess .T his r esult str ongl y points to the antio xidant acti vity of the complexes that, by maintaining the redo x balance at low le v els, no longer r equir es CAT activity to be functional in response to o xidati v e str ess.Se v er al studies r eported that HSP`s plays crucial roles in o xidati v e str ess r esponse, perha ps due to their modulatory effects on inflammation cascades that contr ol ROS gener ation and/or the c ha per one activities assisting misfolded protein to refold properly (Vacher et al. 2005, Hauet-Broere et al. 2006, Grimminger-Marquardt and Lashuel 2010, Morano et al. 2012 ).In this work, the expression of Hsp104 was monitored by the increase in fluorescent cells after treatments, while activation was determined by the formation of foci, c har acterized by the misfolding and a ggr egation of proteins.Indeed, the induction of Hsp104 expression was highly detectable in cells treated with mild heat-shock (40 • C/1 h), and to a lesser extent in cells treated with the complexes.Ho w e v er, foci formation was observ ed onl y in heat-treated cells .T he absence of foci formation in cells treated with the complexes was expected since the exposure of cells to the complexes was non-toxic and, ther efor e, inca pable of inducing protein misfolding and aggregation.These results suggest that the complexes may be triggering a specific cellular signaling pathway that leads to the de novo synthesis of HSP's, essential for adap-tation and additional pr otection a gainst a subsequent stress condition.
Lipid peroxidation caused by oxidative stress conditions, as well as the reduction of this process by antioxidant substances, has been studied by our group (Herdeiro et al. 2006, Horn et al. 2010, Ribeiro et al. 2015 ).The use of antioxidant complexes as an alternativ e to r educe lipid per oxidation has alr ead y been ad dressed in a pr e vious study, whose r esults sho w ed that treating S. cerevisiae with complexes of the HPClNOL series reduced lipid peroxidation v alues by mor e than 50% (Ribeir o et al. 2015 ).Contr asting with r eported data, in which we showed that the Fe(III) compound was slightl y mor e efficient in r educing lipid per oxidation, her ein, no significant changes were observed between the tested complexes.Other antioxidant complexes have also shown a positive effect in r educing lipid per oxidation in differ ent models of study (Gonzalez et al. 1995, Zhang et al. 2004, Clausen et al. 2010 ).The MnSalen deri vati ves EUK-189 and EUK-207 reduced lipid peroxidation by almost 50% in mouse brains subjected to aging for 11 months (Liu et al. 2003 ).Another study sho w ed that treatment with Mn-Salen deri vati ves compounds decreased lipid peroxidation levels in mice fed a diet that caused nonalcoholic steatohepatitis (Rezazadeh et al. 2012 ).
The use of G. mellonella has been r a pidl y disseminated through differ ent ar eas of science (Champion et al. 2016, Fernandes et al. 2017 ).The similarity between the innate immune system of G. mellonella larvae and humans puts this invertebrate in focus as an alternative model for studies on the effect of drugs and fungal/bacterial infection (Kellett et al. 2011, Adamski et al. 2014 , Tr e vijano-Contador and Zar a goza 2018 ).Using Galleria mellonella larv ae we pr ov ed that the complexes act as antioxidants and are safe in terms of their toxicity.We have recently reported the use of G. mellonella larvae to investigate the antioxidant potential of a Mn 2 + -complex, [Mn 2 ( μ 2 -oda)(phen) 4 (H 2 O) 2 ] 2 + , whic h r educed the susceptibility of this inv ertebr ate model to H 2 O 2 str ess (Queir oz et al. 2022 ).Unquestionabl y, r egardless of the treatment dose used, all complexes tested could prolong the survival of G. mellonella larvae.In addition, the complexes also ensured the continuity of the insect's life cycle: all larvae that survived the H 2 O 2 stress reached the adult stage of the life cycle, the moth.This is the first time that monitoring G. mellonella life cycle (i.e. de v elopment fr om larv ae to pupae and then to moth) has been emplo y ed to assess the toxicity, safety of use, and antioxidant activity of coordination compounds.

Conclusions
In this study, we carried out in vitro and in vivo measurements to assess the antioxidant potential of three complexes synthesized with the ligand H 2 BPClNOL.Our r esults r e v ealed that these complexes sho w ed suitable antio xidant acti vities by reducing the susceptibility of S. cerevisiae and G. mellonella to o xidati v e str ess.In this regard, the increased survival observed in S. cerevisiae treated with the synthetic antioxidants seems to be str ongl y r elated to the maintenance of redox homeostasis (e.g.reduction of lipid oxidation and modulation of SOD and CAT activity) and activation of cell str ess r esponse (e.g.induction of HSP104 expr ession).We also observed that S. cerevisiae and G. mellonella well-tolerated these complexes, which in our experimental conditions were non-toxic for both models of stud y.Inclusi ve , all G .mellonella larv ae that wer e treated with the complexes and survived oxidativ e str ess could complete their life cycle, differentiating from larva to pupa and then from pupa to moth.Our results suggest that the complexes may be also triggering a specific cell signaling pathway that culminates in de novo synthesis of pr otectiv e factors suc h as HSP's r equir ed for further protection against a sudden stress condition.Taken together, our results point to a promising antioxidant potential of studied complexes to circumvent the harmful of oxidativ e str ess.
This study sho w ed a new exciting aspect related to the antioxidant activity of coordination compounds by modulating the cellular r esponse, whic h r aises inter est in whether tr eatment with these compounds would be able to atten uate, in alternati ve models of study, the cellular dysfunctions observed in human pathologies.We hope to address such questions in future work.

Figure 1 .
Figure 1.Chemical structure of the complexes 1 (A), 2 (B) and 3 (C) based on the monocrystal x-ray data.

Figure 2 .
Figure 2. In vitro antioxidant activity of 1-3 complexes and [Mn(SALEN)(Cl)].The in vitro e v aluation of radical scavenging activity was determined by the DPPH method.Results were expressed as a percentage of the ability to reduce the DPPH radical and represent the mean ± standard deviation of at least three independent experiments.

Figure 3 .
Figure 3. Complexes 1-3 protect S. cerevisiae to H 2 O 2 stress .T he wild type BY4741 cells, grown in 2% YPD medium, were treated or not for 1 h with differ ent concentr ations of 1 (A), 2 (B) or 3 (C), and then submitted to 1.0 mM H 2 O 2 for1 h.Results ar e expr essed as the percenta ge of surviv al and r epr esent the mean ± standard deviation of at least three independent experiments.* Represents significantly different results compared to the positiv e contr ol (1.0 mM H 2 O 2 ) at P < 0.05.* * Repr esents significantl y differ ent r esults between 25 μM and 50 μM and the other treatments with 1 or 2 at P < 0.05.* * * Represents significantly different results between 12.5 μM, 25 μM and 50 μM and the other treatments with 3 at P < 0.05.

Figure 4 .
Figure 4. Complexes 1-3 protect S. cerevisiae to O 2 r − stress.Cells of the wild type BY4741 str ain, tr eated or not for 1 h with different concentrations of 1 (A), 2 (B) or 3 (C), were subjected to 30 mM menadione/1 h.Cells lacking the Sod1 enzyme ( sod1 ), treated or not for 1 h with different concentrations of 1 (D), 2 (E) or 3 (F), were subjected to 30 mM menadione for 1 h.Results are expressed as the percentage of survival and represent the mean ± standard deviation of at least three independent experiments.* Represents significantly different results compared to the positive control (30 mM Menadione) at P < 0.05.* * Represents significantly different results between 50 μM and 100 μM and the other treatments with 1 or 2 at P < 0.05.* * * Repr esents significantl y differ ent r esults between 12.5 μM, 25 μM, 50 μM and 100 μM and the other treatments with 3 at P < 0.05.

Figure 5 .
Figure 5. Reduction of lipid peroxidation promoted by treatments with complexes 1-3.Cells of the wild type BY4741 str ain, tr eated or not for 1 h with different concentrations of 1, 2 or 3 , were subjected to H 2 O 2 stress, and then prepared for the determination of lipid peroxidation.Results are expressed as pmol of MDA/mg of cell and represent the mean ± standard deviation of at least three independent experiments.* Repr esents significantl y differ ent r esults in r elation to the contr ol ( P < 0.05).* * Represents significantly different results between the treatments and the positive control (1.0 mM H 2 O 2 ) at P < 0.05.* * * Repr esents significantl y differ ent r esults between 50 μM and 25 μM treatments at P < 0.05.

Figure 6 .
Figure 6.Complexes 1-3 mitigate the r equir ement for CAT and SOD activity under oxidative stress.For catalase activity, cell extracts were prepared for enzymatic determination immediately before and after treatments with H 2 O 2 (A).Results were expressed as specific activity ( μmol/min/mg ptn).For SOD activity, exponentially growing cells were subjected to treatment (B), and/or stress with 1 mM of H 2 O 2 .Results were expressed as a fold increase calculated by the ratio of SOD activity between cells treated with different complexes and the control condition.All results represent the mean ± standard deviation of at least three independent experiments.* Represents significantly different results in relation to the control ( P < 0.05).* * Repr esents significantl y differ ent r esults between the tr eatments and the positiv e contr ol (1.0 mM H 2 O 2 ) at P < 0.05.

Figure 7 .
Figure 7. Induction and activation of Hsp104-GFP expression.Cells were treated with 50 μM of the compounds for 1 h or subjected to heat stress (40 • C for 1 h), and then fluorescence was observed under a fluorescence microscope.(A) The percentage of cells expressing Hsp104.(B) The percentage of cells with detectable foci .(C) Re presentati ve fluorescent images of Hsp104-GFP obtained by fluorescent microscopy.Results were expressed as % of cells with Hsp104-GFP expression (A) and as % of cells with Hsp104-GFP foci, r epr esenting the mean ± standard deviation of at least three independent experiments.* Repr esents significantl y differ ent r esults in r elation to the contr ol ( P < 0.05).* * Repr esents significantl y differ ent r esults between the mild heat treatment (40 • C/1 h) and the treatment with 50 μM of the compounds at P < 0.05.

Figure 8 .
Figure 8. Chr onological a ging lifespan extension of BY4741 cells after treatment with the complexes.Cells of the wild type BY4741 strain, treated or not with different concentrations of 1 (A), 2 (B) and 3 (C) complexes, were submitted to chronological aging for 28 da ys .Aliquots were removed every 4 da ys , and the cells were plated with colonies counted after 72 h.Results are expressed as a percentage of survival and represent the mean ± standard deviation of at least three independent experiments.( P < 0.05).* Represents significantly different results in relation to the control ( P < 0.05).

Figure 9 .
Figure 9. Complexes treatment reduce G. mellonella larvae susceptibility to H 2 O 2 stress without interfering in larvae life cycle.Galleria mellonella larvae exposed to treatment with 1 (A), 2 (B) and 3 (C) complexes were subjected to stress with 5 M H 2 O 2 for 1 h.The graphs track larvae survival (9A-C) over a 7-day period and depict the ov er all de v elopment thr oughout the entir e insect life cycle (9D-F).We can observ e the thr ee sta ges: larv ae (D), pupae (E), and moth (F).The gr a phs tr ac k the continuity of the animal's life cycle ov er 21 da ys .Results r epr esent the mean ± standard de viation of at least thr ee independent experiments.